X Examples
of macros
The examples are available interactively in
the VEGA site : tutorials.
We suppose you have created the demo frame files by executing the macros
"create_testfr.C" and the corresponding local database with "create_DB.C" (this
is just a two lines script that does what is described above for creating a
database).
X.1 Example
1 : Displaying a peak
The first example is showing how one can display a potentially interesting
peak, zooming in and plotting two channels on top of each other (macro
peak.C) :
{
// This example shows how to plot a region
of interest.
// To see the result of this macro, click here
// The access of data is made by its
time.
// A much fancier example is given in here
gROOT->Reset();
// Open a new window (canvas)
c1 = new TCanvas("c1","A peak
example",1);
// ===== Creates the needed pads by dividing
the initial canvas =====
c1->Divide(2,2);
// ===== Opens
the database =====
vd = new
VFrDataBase("demoDB.root");
// ===== Gets the needed
vectors and draws them =====
// ===== First the upper plots
=====
FrVect* frv1 =
vd->GetVect("adc.MYGO_ADC1",400000010.,3.);
FrVect*
frvSeis = vd->GetVect("adc.MYGO_Seis",400000010.,3.);
c1->cd(1); // select the first
pad
gVM->Draw(frv1);
c1->cd(2); // select the second
pad
gVM->Draw(frvSeis);
// =====
Gets the needed vectors and draws them =====
// ===== Second
the same plots superimposed =====
// Gets the
needed vectors again but zooming on the region of interest
frv2 = vd->GetVect("adc.MYGO_ADC1",400000010.24,0.06);
frvSeis2 =
vd->GetVect("adc.MYGO_Seis",400000010.24,0.06);
// Draws them in the second pad, superimposed on top of
each other
c1->cd(3);
gVM->Draw(frv2);
gVM->Draw(frvSeis2,"same");
}
To show the result when displaying a the same peak but in a very fancy way
(well, this is all a matter of taste...), with the macro "peak2.C", here is the
result :
Example
2 : scrolling data
This example shows how to display a channel extracted from a database with
scrolling one second at a time. The channel is extracted in the form of FrVect
vectors. Each extracted FrVect does not match the length of the frame it belongs
to. In fact, the user should not care about that.
The example is in the file
scroll.C
{
// This simple example
opens a demo local database and plots a vector.
// The vector
is scrolled 1 s at a time, nvecmax times.
// The extraction
of the vector shows that one can select a vector
// without
having to think about the frame boundaries.
// The length of
the vector is arbitrary and is only limited by the
available
// amount of memory.
gROOT->Reset();
FrVect*
frv;
Double_t time;
// Max number
of vectors displayed
Int_t nvecmax =
20;
// Time step of the scroll
Double_t tistep = 1;
// Length of the displayed
vector
Double_t veclength = 4;
//
Open a new window (canvas)
c1 = new
TCanvas("c1","c1",1);
// Open a
database
vd = new
VFrDataBase("demoDB.root");
// Start
time
time = 400000000.1;
// Get the
first vector
frv =
vd->GetVect("adc.MYGO_ADC1",time,veclength);
Int_t nvec
= 0;
while ( (frv!=0) && (nvec<nvecmax) )
{
nvec++;
printf("nvect = %d,
time = %fn",nvec,time);
// Draws the
vector
gVM->Draw(frv);
// Update
the canvas
gPad->Update();
FrVectFree(frv);
time +=
tistep;
frv =
vd->GetVect("adc.MYGO_ADC1",time,veclength); // Get the next
vector
}
}
X.3 Example
3 : histogramming some data and fitting the result
This example shows how to use histogramming with the VManager. Access is
made to the data and an interesting part containing a peak is displayed, as well
as a histogram of the data. The histogramed data is accumulated over several
frames and, at the last frame, a fit is made to the histogram. One fits a
gaussian to show the deviation from gaussianity.
The example is in the macro scroll2.C
{
// This example plots a vector and
accumulates a histogram of the values
// of this
vector.
// Only the part of the vector that corresponds to
the interesting part
// are plotted (a peak at the beginning
of the frames)
// The histogram is accumulated for nframmax
frames
// To see the result of this macro, click here
-
- gROOT->Reset();
- FrameH* fr;
- Double_t time;
// Max number of
frames displayed
- Int_t nframmax = 14;
// Open a new
window (canvas) and divide it by 2
- c1 = new TCanvas("c1","c1",1);
- c1->Divide(2,1);
- c1_1->SetFillColor(44);
- c1_1->SetFrameFillColor(41);
- c1_2->SetFillColor(44);
- c1_2->SetFrameFillColor(41);
- c1_2->SetLogy();
- c1_2->SetGridx();
- c1_2->SetGridy();
// Open a
database
- vd = new
VFrDataBase("demoDB.root");
// Start
time
- time = 400000000.1;
// Histogram to
constrain the limits
- hisfr = new TH1F("hisfr","Histogram of MYGO_Seis
values",100,0.,3000.);
- hisfr->SetMarkerColor(4);
- hisfr->SetMarkerStyle(8);
- hisfr->SetMarkerSize(0.7);
// Get
the first frame
- fr = vd->GetFrame(time);
- Int_t nfram = 0;
- while ( (fr!=0) && (nfram<nframmax) )
{
- nfram++;
- printf("nfram = %dn",nfram);
//
Draws the vector (only in the signal region)
- c1->cd(1);
-
gVM->Draw(fr,"adc.MYGO_ADC1",0.13,0.07);
//
Get the corresponding VSPlot and set the color to blue
- VSPlot* vfp = gVM->GetLastPlot();
- vfp->SetLineColor(4);
//
Fill the histogram (accumulates it) and draw it
// the +
option in front of the name of the histogram
// means that
it's content is not erased before each fill
- c1->cd(2);
-
gVM->DrawHist(fr,"adc.MYGO_ADC1",0.13,0.07,"+hisfr","e1");
// Update the canvas
- gPad->Update();
- FrameFree(fr);
- fr = vd->GetNextFrame(); // Get the next
frame
- }
// Fits the histogram and draws
the fit result
- hisfr->Fit("gaus");
- hisfr->SetMarkerColor(4);
- hisfr->SetMarkerStyle(8);
- hisfr->SetMarkerSize(0.7);
-
hisfr->Draw("e1");
}
X.4 Example
4 : Extracting some slow monitoring data and doing some simple plots
This example shows how to extract slow monitoring data and plot variables
or combinations of them.
The example is in the macro
sms.C
{
// This example shows how
to plot a set of slow monitoring values.
// To see the result
of this macro, click begin_html <a href="gif/sms.gif">here</a>
end_html
// A ntuple is build and used to plot various
combinations of variables.
gROOT->Reset();
// Open a new window (canvas)
c1 = new TCanvas("c1","A slow monitoring
example",1);
// ===== Creates the needed pads by dividing
the initial canvas =====
c1->Divide(1,2);
// ===== Opens
the database =====
vdsms = new
VFrDataBase("SMSdemoDB.root");
// ===== First builds the
ntuple from the MYGOSMS station values =====
nt =
vdsms->ExtractSMS("nt","MYGOSMS",400000000,5000);
//
===== Then draws the plots =====
c1->cd(1);
// select the first pad
// Sets some style
options
gStyle->SetMarkerColor(50);
gStyle->SetMarkerStyle(8);
gStyle->SetMarkerSize(0.5);
// Draws the value VF1
with respect to time
// t is automatically filled as the
first variable of the ntuple at extraction
nt->DrawSeries("MYGOSMS.VF1","","P");
vs =
gVM->GetLastPlot();
vs->SetTitle("");
// Sets some
style options
gStyle->SetLineColor(4);
// Superimposes the values
of sqrt(VI2+50) (it's an example)
nt->DrawSeries("sqrt(MYGOSMS.VI2+50)","","same");
// Draws a title for the graphs
pl =
new TPaveLabel(-400,135,1000,156,"VF1,sqrt(VI2+50)","br");
pl->Draw();
// ===== In the second
pad, draws a histogram =====
// ===== of values VF1 with
respect to sqrt(VI2+50) =====
c1->cd(2);
// select the second pad
// Draws the
histogram
nt->Draw("MYGOSMS.VF1:sqrt(MYGOSMS.VI2+50)","","cont0");
}
Here is the result of this macro :
X.5 Example
5 : Extracting a vector and plotting spectra
This example shows how to use the fft algorithms to plot spectra.
The example is in the macro scrollspectrum.C
{
// This simple example opens a demo
metadatabase,
// plots a vector and it's
spectrum.
// The vector is scrolled 1 s at a time, nvecmax
times.
// The extraction of the vector shows that one can
select a vector
// without having to think about the frame
boundaries.
// The length of the vector is arbitrary and is
only limited by the available
// amount of
memory.
// It is also shown how to use the FFT from the
signal analysis library
gROOT->Reset();
gBenchmark->Start("scrollspectrum");
FrVect* frv;
// Max number of vectors
displayed
Int_t nvecmax = 20;
// Time step of the
scroll
Double_t tistep = 1;
// Length of the displayed
vector
Double_t veclength = 4;
// Start
time
Double_t time = 400000000.1;
// Open a new window
(canvas) and split it
c1 = new TCanvas("c1","c1",1);
c1->Divide(1,2);
c1->cd(2);
// Open a database
vd = new VFrDataBase("demoDB.root");
// Get the first
vector
frv = vd->GetVect("adc.MYGO_ADC1",time,veclength);
Int_t nvec = 0;
while ( (frv!=0) && (nvec<nvecmax) ) {
nvec++;
printf("nvect = %d, time = %f\n",nvec,time);
//
Draws the vector in the first pad
c1->cd(1); // go to the first pad
gVM->Draw(frv);
// Calculates the
spectrum
FrvRFFT* fft;
fft = FrvRFFTFor(0,frv);
// Draws the spectrum
in the second pad
c1->cd(2); // go to the second pad
gVM->Draw(fft->amplitude);
gPad->SetLogx(1);
gPad->SetLogy(1);
// Update the
canvas
c1->cd();
gPad->Update();
// Release the various
variables allocated
FrVectFree(frv);
FrvRFFTFree(fft);
time += tistep;
frv = vd->GetVect("adc.MYGO_ADC1",time,veclength); // Get
the next vector
}
gBenchmark->Show("scrollspectrum");
// Do some garbage collection
if (frv) FrVectFree(frv);
delete vd;
}
Damir BUSKULIC
Last update :19/11/2001;
